From: Dan Williams <dan.j.williams@intel.com>
To: Al Viro <viro@zeniv.linux.org.uk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
linux-arch@vger.kernel.org, Andi Kleen <ak@linux.intel.com>,
Kees Cook <keescook@chromium.org>,
kernel-hardening@lists.openwall.com,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
"the arch/x86 maintainers" <x86@kernel.org>,
Ingo Molnar <mingo@redhat.com>, "H. Peter Anvin" <hpa@zytor.com>,
Thomas Gleixner <tglx@linutronix.de>,
Andrew Morton <akpm@linux-foundation.org>,
Alan Cox <alan@linux.intel.com>
Subject: Re: [PATCH v3 8/9] x86: use __uaccess_begin_nospec and ASM_IFENCE in get_user paths
Date: Tue, 16 Jan 2018 22:50:01 -0800 [thread overview]
Message-ID: <CAPcyv4jn6xHNB2DevoFEzkfpnKCd7u8UdA+TmVPtbO08TjzFWA@mail.gmail.com> (raw)
In-Reply-To: <20180117062800.GU13338@ZenIV.linux.org.uk>
On Tue, Jan 16, 2018 at 10:28 PM, Al Viro <viro@zeniv.linux.org.uk> wrote:
> On Tue, Jan 16, 2018 at 08:30:17PM -0800, Dan Williams wrote:
>> On Tue, Jan 16, 2018 at 2:23 PM, Dan Williams <dan.j.williams@intel.com> wrote:
>> > On Sat, Jan 13, 2018 at 11:33 AM, Linus Torvalds
>> [..]
>> > I'll respin this set along those lines, and drop the ifence bits.
>>
>> So now I'm not so sure. Yes, get_user_{1,2,4,8} can mask the pointer
>> with the address limit result, but this doesn't work for the
>> access_ok() + __get_user() case. We can either change the access_ok()
>> calling convention to return a properly masked pointer to be used in
>> subsequent calls to __get_user(), or go with lfence on every
>> __get_user call. There seem to be several drivers that open code
>> copy_from_user() with __get_user loops, so the 'fence every
>> __get_user' approach might have noticeable overhead. On the other hand
>> the access_ok conversion, while it could be scripted with coccinelle,
>> is ~300 sites (VERIFY_READ), if you're concerned about having
>> something small to merge for 4.15.
>>
>> I think the access_ok() conversion to return a speculation sanitized
>> pointer or NULL is the way to go unless I'm missing something simpler.
>> Other ideas?
>
> What masked pointer?
The pointer value that is masked under speculation.
diff --git a/arch/x86/lib/getuser.S b/arch/x86/lib/getuser.S
index c97d935a29e8..4c378b485399 100644
--- a/arch/x86/lib/getuser.S
+++ b/arch/x86/lib/getuser.S
@@ -40,6 +40,8 @@ ENTRY(__get_user_1)
mov PER_CPU_VAR(current_task), %_ASM_DX
cmp TASK_addr_limit(%_ASM_DX),%_ASM_AX
jae bad_get_user
+ sbb %_ASM_DX,%_ASM_DX
+ and %_ASM_DX,%_ASM_AX
ASM_STAC
1: movzbl (%_ASM_AX),%edx
xor %eax,%eax
...i.e %_ASM_AX is guaranteed to be zero if userspace tries to cause
speculation with an address above the limit. The proposal is make
access_ok do that same masking so we never speculate on pointers from
userspace aimed at kernel memory.
> access_ok() exists for other architectures as well,
I'd modify those as well...
> and the fewer callers remain outside of arch/*, the better.
>
> Anything that open-codes copy_from_user() that way is *ALREADY* fucked if
> it cares about the overhead - recent x86 boxen will have slowdown from
> hell on stac()/clac() pairs. Anything like that on a hot path is already
> deep in trouble and needs to be found and fixed. What drivers would those
> be?
So I took a closer look and the pattern is not copy_from_user it's
more like __get_user + write-to-hardware loops. If the performance is
already expected to be bad for those then perhaps an lfence each loop
iteration won't be much worse. It's still a waste because the lfence
is only needed once after the access_ok.
> We don't have that many __get_user() users left outside of arch/*
> anymore...
next prev parent reply other threads:[~2018-01-17 6:50 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-01-13 18:17 [PATCH v3 0/9] core, x86: prevent bounds-check bypass via speculative execution Dan Williams
2018-01-13 18:17 ` [PATCH v3 1/9] Documentation: document array_ptr Dan Williams
2018-01-13 18:17 ` [PATCH v3 2/9] arm64: implement ifence_array_ptr() Dan Williams
2018-01-13 18:17 ` [PATCH v3 3/9] arm: " Dan Williams
2018-01-13 18:17 ` [PATCH v3 4/9] x86: implement ifence() Dan Williams
2018-01-13 18:17 ` [PATCH v3 5/9] x86: implement ifence_array_ptr() and array_ptr_mask() Dan Williams
2018-01-13 18:17 ` [PATCH v3 6/9] asm/nospec: mask speculative execution flows Dan Williams
2018-01-13 18:18 ` [PATCH v3 7/9] x86: introduce __uaccess_begin_nospec and ASM_IFENCE Dan Williams
2018-01-13 18:18 ` [PATCH v3 8/9] x86: use __uaccess_begin_nospec and ASM_IFENCE in get_user paths Dan Williams
2018-01-13 19:05 ` Linus Torvalds
2018-01-13 19:33 ` Linus Torvalds
2018-01-13 20:22 ` Eric W. Biederman
2018-01-16 22:23 ` Dan Williams
[not found] ` <CA+55aFxAFG5czVmCyhYMyHmXLNJ7pcXxWzusjZvLRh_qTGHj6Q@mail.gmail.com>
2018-01-16 22:41 ` Linus Torvalds
2018-01-17 14:17 ` Alan Cox
2018-01-17 18:52 ` Al Viro
2018-01-17 19:54 ` Dan Williams
2018-01-17 20:05 ` Al Viro
2018-01-17 20:14 ` Dan Williams
2018-01-18 3:06 ` [RFC][PATCH] get rid of the use of set_fs() (by way of kernel_recvmsg()) in sunrpc Al Viro
2018-01-18 3:16 ` Linus Torvalds
2018-01-18 4:43 ` Al Viro
2018-01-18 16:29 ` Christoph Hellwig
2018-01-18 17:10 ` Al Viro
2018-01-18 19:31 ` Al Viro
2018-01-18 19:37 ` [PATCH 01/10] net: separate SIOCGIFCONF handling from dev_ioctl() Al Viro
2018-01-18 19:37 ` [PATCH 02/10] devinet_ioctl(): take copyin/copyout to caller Al Viro
2018-01-22 16:40 ` Christoph Hellwig
2018-01-18 19:37 ` [PATCH 03/10] ip_rt_ioctl(): take copyin " Al Viro
2018-01-22 16:43 ` Christoph Hellwig
2018-01-18 19:37 ` [PATCH 04/10] kill dev_ifsioc() Al Viro
2018-01-22 16:47 ` Christoph Hellwig
2018-01-18 19:37 ` [PATCH 05/10] kill bond_ioctl() Al Viro
2018-01-22 16:48 ` Christoph Hellwig
2018-01-18 19:37 ` [PATCH 06/10] kill dev_ifname32() Al Viro
2018-01-18 19:37 ` [PATCH 07/10] lift handling of SIOCIW... out of dev_ioctl() Al Viro
2018-01-18 19:37 ` [PATCH 08/10] ipconfig: use dev_set_mtu() Al Viro
2018-01-18 19:37 ` [PATCH 09/10] dev_ioctl(): move copyin/copyout to callers Al Viro
2018-01-18 19:37 ` [PATCH 10/10] kill kernel_sock_ioctl() Al Viro
2018-01-22 16:49 ` Christoph Hellwig
2018-01-24 20:52 ` David Miller
2018-01-25 0:01 ` Al Viro
2018-01-25 0:21 ` Al Viro
2018-01-25 4:11 ` David Miller
2018-01-22 16:40 ` [PATCH 01/10] net: separate SIOCGIFCONF handling from dev_ioctl() Christoph Hellwig
2018-01-18 20:33 ` [RFC][PATCH] get rid of the use of set_fs() (by way of kernel_recvmsg()) in sunrpc Al Viro
2018-01-19 3:27 ` Al Viro
2018-01-17 19:26 ` [PATCH v3 8/9] x86: use __uaccess_begin_nospec and ASM_IFENCE in get_user paths Linus Torvalds
2018-01-17 20:01 ` Eric Dumazet
2018-01-18 16:38 ` Christoph Hellwig
2018-01-18 16:49 ` Linus Torvalds
2018-01-18 18:12 ` Al Viro
2018-01-17 4:30 ` Dan Williams
2018-01-17 6:28 ` Al Viro
2018-01-17 6:50 ` Dan Williams [this message]
2018-01-17 10:07 ` David Laight
2018-01-17 18:12 ` Dan Williams
2018-01-17 19:16 ` Linus Torvalds
2018-01-13 18:18 ` [PATCH v3 9/9] vfs, fdtable: prevent bounds-check bypass via speculative execution Dan Williams
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAPcyv4jn6xHNB2DevoFEzkfpnKCd7u8UdA+TmVPtbO08TjzFWA@mail.gmail.com \
--to=dan.j.williams@intel.com \
--cc=ak@linux.intel.com \
--cc=akpm@linux-foundation.org \
--cc=alan@linux.intel.com \
--cc=gregkh@linuxfoundation.org \
--cc=hpa@zytor.com \
--cc=keescook@chromium.org \
--cc=kernel-hardening@lists.openwall.com \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=viro@zeniv.linux.org.uk \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).